See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/254050489

Non-linear load compensation in Fuel Cell grid interfaced system using active
power filter

Article · December 2011

DOI: 10.1109/PEDS.2011.6147246

CITATIONS READS
15 100

3 authors, including:

Gitanjali Mehta S P Singh

Galgotias University, Greater Noida, India Jawaharlal Nehru Krishi Vishwavidyalaya
46 PUBLICATIONS 306 CITATIONS 82 PUBLICATIONS 1,584 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Gitanjali Mehta on 30 March 2017.

The user has requested enhancement of the downloaded file.

 IEEE PEDS 2011, Singapore, 5 - 8 December 2011
1

Non-linear load compensation in Fuel Cell grid

interfaced system using active power filter
Gitanjali Mehta1, Member IEEE and S. P. Singh2, Member IEEE, R. D. Patidar3
1,2
Electrical Engineering Department, Indian Institue of Technology, Roorkee, India
3
Mandsaur Institute of Technology, Mandsaur, India

Abstract - The increasing global warming concerns and increased rapidly in the last two decades. Many APF
diminishing fossil fuels have made us necessary to look for topologies and control algorithm are reported in literature. But
alternative sources of energy. Fuel Cell technology holds promise the installation of the APF requires additional costs. The cost
towards sustainable power generation, it being pollution free and of installing FC plant is also large. Thus, the function of APF
using readily available fuels. This paper presents the modeling,
can be implemented with the FCDG system thus reducing the
control and design analysis of a three-phase Grid-interactive Fuel
Cell system with active filter functions. The main focus of this overall cost. Further, the cost of the Grid-interactive FC
paper is to control the active power supplied by the Fuel Cell system can be minimized by decreasing the number of power
Distributed Generation system while compensating harmonics conversion stages and the number of components involved in
and reactive currents caused by the nonlinear loads using shunt each stage.
active power filter. The developed Fuel Cell model is connected to Thus, to reduce the cost and increase the efficiency, a
the DC-side of the voltage source inverter for interfacing with the single-stage, three-phase, grid-interactive FCDG system with
grid. Thus the same inverter is utilized as power converter to the functionality of APF has been proposed in this paper. The
inject the power generated from the Fuel Cell to the grid and to function of APF is added in the FC system inverter by making
act as active power filter to compensate load current harmonics
the necessary modification in the control methodology. The
and load reactive power demand. The designed controller either
regulates the power flow between the Fuel Cell and the Grid or proposed interface does not require any additional circuit for
works as an active power filter or performs both the functions enabling the existing inverter to also perform the APF
simultaneously. The simulation model of the overall system is function. This concept reduces the overall design cost of the
developed in MATLAB/Simulink environment using SimPower system. The proposed system capable of injecting FC power to
Systems blocksets and then PIL simulated using TMS320F2812 electric grid also compensates load reactive power and
DSP. The results are obtained for different operating conditions harmonics caused by non-linear loads.
with varying load demands to prove the effectiveness of the entire
system. II. MODELING AND CHARACTERISTICS OF FC STACK
Index Terms - Fuel Cell, Distributed Generation, Active, The FCs are connected in series to give sufficient DC
Reactive and Harmonic Power, Active Power Filter voltage, it results in a FC stack. The FC stack can be modeled
as a controlled voltage source E in series with a constant
I. INTRODUCTION resistance R as shown in Fig. 1. The controlled voltage
source can be represented a
T he global energy consumption is increasing at a rapid
pace. However, the conventional fossil fuels, the primary
source of electric power are on the verge of extinction [1].
E E NA ln
and FC stack output voltage as:
(1)

Also, the extensive use of fossil fuels and nuclear resources V E R i (2)
cause serious environment pollution and safety problems [2]. where E is the open circuit voltage, N is the number of cells
Distributed generation (DG) system is rapidly increasing in the stack, A is the Tafel slope, i is the exchange current
around the world because they can meet both the increasing resulting from the continual backward and forward flow of
power demand and environmental regulations of low electrons from and to the electrolyte at no load, R is the
emissions. Fuel cell (FC) based power generation is gaining internal resistance of the FC stack, i is the FC stack current
popularity due to its cleanliness, portability and suitability for and V is the FC stack voltage. The Tafel Slope, exchange
electricity and heat generation. FC is an electrochemical current and the open circuit voltage are affected by variations
device that converts the chemical energy of hydrogen fuel in temperature and partial pressures of hydrogen and oxygen.
directly into electrical energy [3]. It offers various advantages They can be represented as:
such as high energy conversion efficiency, modular design, E KE (3)
and low environmental intrusion as compared to conventional F H O ∆G
i exp (4)
power generators [4, 5]. R RT
RT
The increased use of power electronics devices in various A (5)
F
loads result in many power quality (PQ) problems in the ac where E is the Nernst voltage, the thermodynamics voltage of
network of which the harmonics and reactive power are of the cells and depends on the temperature and partial pressures
major concern. This may lead to heating and malfunctioning of reactants inside the stack and is given as:
of sensitive devices. In order to solve the aforesaid PQ . RT
problems, the use of active power filter (APF) systems have E 0.229 T 298 ln pH pO (6)
F F

978-1-4577-0001-9/11/$26.00 ©2011 IEEE 197

 2

where α is the charge transfer coefficient, w which depends on V V NA I

R (11)
the type of electrodes and catalysts used. T is the operating I
V E R
temperature of the cell, pH is the partial presssure of hydrogen i exp (12)
NA
inside the stack, pO is the partial pressure of oxygen inside the Thus from the four points on the pollarisation curve, we derive
stack, k is the Boltzmann’s constant (1.38 10 J⁄K), h is the values of i , R and NA. The material
m parameters K, ∆G
the Plank’s constant (6.626 10 Js), ∆G
G is the size of the and α are calculated based on the t polarization curve at
activation barrier which depends on the typee of electrode and nominal conditions of operation along with some additional
catalyst used, z is the number of moving elecctrons and K is the parameters, such as the low heatinng value efficiency of the
voltage constant at nominal operating condition. stack, composition of fuel and air,
a supply pressures and
temperatures etc [5, 6].
⎛ i fc ⎞
E = E oc − NA ln⎜⎜ ⎟⎟
III. ACTIVE POWER FILTER
⎝ i0 ⎠
i fc
R fc Fig. 3 shows the basic compensatiion principle of the shunt
active power filter (SAPF). The sup pply voltage can be written
i fc
+ as:
v t V sinωt (13)
+
E V fc where V is the peak value of sourcce voltage. The non-linear
−
load will draw current that will have a fundamental component
_
and harmonic components which can n be represented as:
iL t ∑∞ I sin nωt φ I sin ωt φ
∑ I sin nωt φ (14)
Fig. 1. Fuel Cell Stack Modeel where I and I are the peak vallues of fundamental load
current and n harmonic load curren nts respectively. and
The modeling of PEMFC is based on the eqquations ((1)-(6)).
are the phase angles of the fun ndamental and harmonic
The experimental measured voltagee-current (V-I)
component of the load current respeectively. The instantaneous
characteristic of NedStack 100 kW PEMFC C Stack is used to
power drawn by load can be given ass:
derive the Matlab/Simulink model of the FC stack. Fig. 2
pL t v t iL t
shows the V-I and P-I characteristics of the FC Stack.
V I sin ωt cosφ
V I sin ωt cos ωt sin φ

V sin ωt I sin nωt φ

p t p t p t (15)

Rs Ls iL
is vs

ic
Grid
Rc Non-linear
Shunt load
Active
Power Lc
Fig. 2. The V-I and P-I characteristic off FC stack Filter

To determine the parameters E , i , NA annd R , four points Vdc

are taken on the V-I curve. These points corrrespond to current C dc
and voltage at nominal operating point (I ,V , current
and voltage at maximum operating point (I , V and
voltage at 0A current E and voltage at 1A current (V .
The FC Stack parameters extracted from m the Nedstack
iL
datasheet V-I curve is given in Table 1 in Apppendix.
The following relations are derived from equaations (1) and (2):
V E NA ln i R (7) is
I
V E NA ln R I (8)
I ic
V E NA ln R I (9)
Solving equations ((7)-(9)), we get
V V I V V I
NA (10)
I I I I Fig. 3. Compensation principle of Shunt
S Active Power filter

198
 3

The first term is the instantaneous load active power, the Fig. 5 represents the vector diagram at unity power factor
second term is the instantaneous load reactive power and the mode of operation. In this figure V and V are the rms
third term is the instantaneous load harmonic power. A SAPF voltages at PCC and AC-side of the inverter, respectively. Is,
is required to be connected in parallel with the non-linear load IL and Ic are the rms values of currents to the grid, to load and
to detect its reactive and harmonic current and to inject into from the inverter respectively. Isp and Isq are the active and
the system a compensating current equal to the load reactive reactive components of grid current Is, ILp and ILq are the active
and harmonic current. The grid needs to supply only the real and reactive components of load current IL while Icp and Icq
power requirement of the load, the reactive and harmonic are the active and reactive components of inverter output
power is to be provided by the compensator. The SAPF current Ic, respectively. ΦL is the load power factor angle. δc is
consists of an inductance L and a resistance R per phase on the angle between Vc and Vs.
the AC-side of three-phase IGBT bridge current controlled-
voltage source inverter (CC-VSI) with a DC-link capacitor C B
[7-9]. Ic Vc
I cq X c Ic
IV. PROPOSED FCDG SYSTEM WITH POWER FLOW AND POWER
QUALITY CONTROL Vcs
δc
I sp I Lp I fc Ic R
A. Power Circuit Description O Vs A
φL
The single-line diagram of proposed three-phase Grid-
interactive FC system is shown in Fig. 4. A current controlled I Lq
three-phase SAPF with energy storage capacitor is IL
connected in parallel with non-linear loads. The purpose of the
SAPF is to maintain the source currents to be sinusoidal, and Fig 5. Vector diagram at unity power factor operation mode
in-phase with the fundamental component of source voltages
in spite of non-linear load connected at the point of common Taking R very small as compared to X , current I through
coupling (PCC). A FCDG unit is connected to the DC-link of APF inductor can be written as
APF. Thus in the proposed FC system, the APF can V
I (18)
compensate the current harmonics and load power factor while X
the FCDG supplies power to the grid and load. From the vector diagram of Fig. 5, we get
V V V 2V V cos δ (19)
V fc
Vc ∠δ c
R L Vs ∠0 From Equations (11) and (12), we get
Ic Is
DC V V V V
FCDG
AC
Grid Hence, I (20)
X
Cdc IL The active power P and reactive power Q flow from FCDG to
Non-linear the PCC can be given by
load V V V V
P sin δ sin δ (21)
X X
V V
Q V cos δ V m V cos δ V (22)
X X
Fig. 4. Single line diagram of Proposed Grid-interactive Fuel
The real and reactive power flow from the FC to the grid and
Cell system
APF reference current is a function of magnitude of
B. Control Theory Analysis the VSI output and phase angle (δc ) between Vc and Vs. and
thus be controlled by regulating the inverter output voltage
To examine the overall control circuit of FCDG system lets both in magnitude and phase.
assume the load current consists of a set of harmonic C. Control Circuit Description
components h, where h 1, 2 … ∞ . The PCC voltage vector
v t and load current vector iL t can be expressed as: v sa sin(ωt )
Phase
v V sin ωt v sb Lock
Unit vector
calculation
sin(ωt − 120°) +
−
v sc Loop sin(ωt + 120°) Gate pulses
v t v V sin ωt 120° (16) + to the
− inverter
v V sin ωt 120° Pfc
+
iL ∑h ILh sin hωt h Load Active
−
i La PL Ps I s*1 Hystersis Current
iL t iL ∑h ILh sin h ωt 120° h (17) i Lb
power
calculation
+
−
×
÷ i La i Lb i Lc Controller

iL ∑ ILh sin h ωt 120° h i Lc Pl V s1

V ,V ,V are the peak values of PCC voltages,

* Voltage
ILh , ILh , ILh are the peak values of load currents Vdc +
− regulator

corresponding to h h order harmonics, and h , h , h are Vdc

the phase angles. Fig. 6. Control circuit of Grid interfacing inverter

199
 4

The control circuit is developed as per Fig 6. For Yout of model

compensation at unity power factor, the active power supplied TMS320F2812
MATLAB
by the source is equal to the source apparent power as given Simulink Rate Limiter Digital Signal
Processor
by:
P PL P P V I (23) Yout of controller
where PL is the load active power, P is the active loss power
Fig. 7. Block diagram of Processor-in-the loop Simulation
of the inverter and P is the active power supplied by the FC
stack, V is the peak value of the PCC voltages and I is the At each time step, MATLAB simulates the plant model for
peak value of the desired source currents after compensation. one sample interval and exports the system output to the
Thus processor. When the processor receives signals from the plant
P
I (24) model, it executes the control code for one sample interval.
V
The reference instantaneous source currents i , i , i are The processor returns the control signals computed during this
computed using the desired peak value and unit current step to MATLAB, via the same link. Thus one cycle of the
vectors u , u , u derived from sensed source voltages simulation completes and the plant model proceeds to the next
using phase-locked-loop (PLL). sample interval. Fig. 8 shows the PIL simulated model of the
proposed system. Fig. 9 shows the control circuit of the
i t I sin ωt
proposed system which is created in PIL subsystem for its real
i t I sin ωt 120° (25)
time simulation.
i t I sin ωt 120°
The compensating current is computed by taking the
difference between instantaneous source reference currents Ic_abc [Ic_abc]

and sensed load currents.

P_f c [P_fc]
PQf c
[Pfuel ]
[Pload] In1 a
FC_1

i t i t iL t
FC_2
V_a v _s [Vs_abc]
G V_a [Pfuel] In2 b

FC_1

i t i t iL t
FC_2

(26)
[G] [Pgrid] In3 c
V_b i_s [Is_abc]
V_b

PEM Fuel Cel l system

i t i t iL t
V_c

DC-AC Converter V-c PQ [Pgrid]

V_a

V_b

V_c
The compensating currents are processed through hysteresis Load
Gri d Source
[Vs_abc]

controllers to provide switching pulses to the IGBTs of the

IL_abc
t

PQ
[Is_abc]
Clock
To Workspace

PWM inverter. The output equation of hysteresis controller for

[Pload]
Discrete, [IL_abc]
Ts = 1e-005 s.
[IL_abc]

phase-a is given as: [Ic_abc]

1 if ∆i
S_g1
S_g1 V_abc
Convert [Vs_abc] [Vdc]

S hys i t ∆i (27)
S_g2
S_g2

0 if ∆i
IL_abc
S_g3 Convert [IL_abc]
S_g3
[G] PIL
S_g4

where, b is the hysteresis band and Sw is the status of the

S_g4 Vdc
Convert [Vdc]
S_g5
S_g5

switches. The output of hysteresis controller regulates the duty

Vq_ref
S_g6 Convert [Pfuel]
S_g6

cycle of the PWM inverter. The switching status of one switch

PIL TMS 320C2000

in a leg is complementary to another, and hence, when S1 = 1,

S4 = 0 or vice versa. Similar pattern of pulses are also Fig. 8. Overall PIL co-simulation model of the proposed
generated for other phases. grid-interactive Fuel Cell system

D. Processor-in-the-loop simulation
1
2
S _g1

The Processor-in-the-loop (PIL) simulation is used to test

Pfuel In 1
1 In1 2
In1 Is1* isa*
IL _abc Ps Out1 S_g2
Pload Out1 In1
Out1 Out1

the control algorithm in the Digital Signal Processor (DSP). A

In 2 In 1 ica *
Out1 Out1 In1 3
In2 isb* Out2
In2 S_g3
In2 In 3 Out2 In 3
Ploss Out3 4

simulation model is developed in Matlab/Simulink and then

Peak source current reference In3 icb *
isc* Out2 In2 S_g4
3 Grid power calculation Out4
Load active power calculation In4 Out3 In 5
Vs_abc
icc* 5

emulated using eZdsp F2812 DSP board. In PIL simulation,

Out5
Reference source current Out3 In3
S_g5
In 6
Out6 6

the system model remains in Matlab while the control

Comparator S_g6
4 In1 Out1 Hysterisis current controller
Vdc

subsystem is implemented in the processor hardware. The Loss power calculation

Simulink model interacts with the DSP via a serial connection. Out2 In1 Out1
ua *
F2812 eZdsp

The softwares and hardwares used are Matlab R2008b with Out3 Out2
ub *
In1 In2

uc *

Embedded Target for TI C2000, Code Composer Studio 3v3,

Out4 In3 Out3

Positive sequence detector Unit vector calculator

and Spectrum Digital eZdsp F2812.

Code Composer Studio™ (CCStudio) is the integrated Fig. 9. PIL control of proposed grid-interactive Fuel Cell
development environment for TI's DSPs, microcontrollers and system
application processors. CCStudio includes a set of tools to
develop and debug embedded functions. Rate limiter and A/D V. RESULTS AND DISCUSSIONS
converters are required to convert the output signals from the
Simulink model into DSP compatible signals. Fig. 7 gives the The parameters used for PIL simulation is given in Table 2 in
overview of the emulation process [9]. Appendix. The three-phase uncontrolled rectifier block with
an R-L branch load on its DC-side is used as non-linear load

200
 5

connected at the PCC. The inverter connectting the FCDG to

the grid is switched on at t=0.1s. The FC stacck capacity is 100
kW. The FCDG supply active and reactive power to local
loads as per the load requirement and rest off the active power
is injected to the grid. If the load active poweer demand is more
than the FC stack capacity, both FC and ggrid supply active
power to the load.
The PCC voltage v , grid current i , loaad current iL and
inverter output current i before and after switching on the
inverter is shown in Fig. 10. After switching on the inverter at
t=0.1s, the grid current become pure sinusoidal and in phase
with the grid voltage. For this period thee reactive power
supplied by the grid is almost zero.

Fig 12. Simulated response of Grid, Fuel Cell Stack and Load
Power.

The total harmonic distortion (THDD) of the grid currents is

measured. A sample of FFT of the grid
g current, phase-a before
connecting the FCDG is shown in Fiig. 13 and after connecting
the FCDG is shown in Fig. 14. The THD of current is reduced
from 17.66% to 3.47%.
Fig. 10. Grid voltage Vs (V), Grid current is ((A), Load Current
iL (A) and Inverter output current iv (A)

Fig. 11 shows the three phase grid voltages ( scaled by a factor

of 0.5) and grid currents. Initially the entiree load demand is
provided by the grid. At t=0.1s, the entire loaad power demand
is provided by the FC and the remaining poower supplied by
FC is given to grid. It is depicted from the ffigure that during
the operation of FC/APF inverter (t > 0.1 s) grid current is in-
phase with the respective voltage and perfectly sinusoidal in
nature. This proves the harmonic as welll as the reactive
current compensation capability of the proposed Grid-
interactive FC system. Fig 13. A sample of Grid current annd its harmonic spectrum
for t < 0.1ss
Vsa (V) and Isa (A)

500

-500
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Vsb (V) and Isb (V)

500

-500
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Vsc (V) and Isc (A)

500

-500
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Time (s)

Fig 11. Grid Voltage and Curreent.

Fig. 12 shows the load active power, FC stacck power and grid
active power. The initial load active power ddemand is 50 kW.
Hence the FC stack provides 50 kW power to the local load
and the remaining 50 kW power is supplieed to the grid. At Fig 14. A sample of Grid current annd its harmonic spectrum
t=0.5s, the load active power demand is incrreased to 175 kW. for t > 0.1ss
Since the FC stack capacity is less than the lload active power
requirement, both FC and grid supply active ppower to the local VI. CONCLUSSION
load. The load active power demand is chaanged back to 50
kW at t=0.8s. This paper presents the modeling, deesign and analysis of Grid-
interactive Fuel Cell system with h embedded active filter
functions under conditions of varrying load demand. The
simulation model is developed in Matlab/Simulink software

201
 6

and then co-simulated in TMS320F2812 digital signal to three-phase active filter for harmonics and reactive power
compensation,” IEEE Transactionss on Power Systems, vol. 13, no.
processor through processor in loop usingg CCStudio. The
1, February 1998, pp. 133-138
results show that the voltage source inverterr not only injects [8] M. Dai, M. N. Marwali, J. W. Jung, and A. Keyhani, “Power Flow
active power at the point of common cooupling but also Control of a Single Distributed d Generation Unit,” IEEE
compensates load reactive power and mainttains a sinusoidal Transactions on Power Electronicss, vol. 23, no. 1, pp. 343-352, Jan.
2008.
current from/to grid. Thus a comm mon inverter is
[9] S. Rahmani, N. Mendalek and K. Al-Haddad, “Experimental
simultaneously utilized to inject power geneerated to the grid Design of a Nonlinear Control Technique for Three-Phase Shunt
and to improve the power quality at pooint of common Active Power Filter,” IEEE Transaactions on Industrial Electronics,
coupling. The results also confirm the good dynamic vol. 57, no. 10, October 2010, pp. 3364-3375.
3
[10] M. Singh, V. Khadkikar, A. Chandra and R. K. Varma, “Grid
performance of the SAPF for change in non-llinear loads.
Interconnection of Renewable Energy Sources at the Distribution
Level With Power-Quality Im mprovement Features,” IEEE
VII. APPENDIX Transactions on Power Delivery, vol.
v 26, no. 01, January 2011, pp.
307-315.
[11] A. Chaoui, J. P. Gaubert, F. Krim K and G. Champenois, “PI
Table 1 Fuel Cell Stack parameeters
controlled three-phase shunt activve power filter for power quality
improvement,” International Journal of Electric Power Components
Parameters Values and Systems, vol. 35, no. 12, Dec. 2007, pp. 1331-1344.
Voltage at 0 A = 890 V
Voltage at 1A = 885 V
Current at nominal operating point = 145.3 A Gitanjali Mehta was
w born in Allahabad, UP, India
Voltage at nominal operating point = 600 V in 1980. She recceived her B.Tech degree in
Electronics and Coommunication Engineering from
Current at maximum operating point = 280 A M. J. P. Rohilkhannd University, Bareilly, India in
Voltage at maximum operating point = 440 V 2003 and M.Tech degree
d in Power Electronics and
ASIC Design from M.N.N.I.T. Allahabad, India in
2005. Currently shee is pursuing the Ph.D. degree in
mulation
Table 2 Parameters used for PIL sim the field of Diistributed Generation in the
Department of Electrical
E Engineering, Indian
Parameters Values Institute of Technology, Roorkee, India. Her research interest includes
FC Stack Power 100 kW renewable and distributed generation sy ystems and power quality control
using active filters.
Grid Voltage and 440 V Ph-Ph (Vrmms), 50 Hz
Frequency Sajjan Pal Singh h was born in Aligarh, UP, India
Grid Impedance per R=0.01Ω, L=1mH
H in 1956. He receivved his B.Sc degree in Electrical
Engineering from m Aligarh Muslim University,
phase Aligarh, India in the year 1978. He received his
Filter resistance and R=0.1 Ω, L=1mH
H M.E. and Ph.D. degrees
d in Electrical Engineering
inductance per phase from Indian Insttitute of Technology, Roorkee
DC-link capacitance 1000uF (formerly Universsity of Roorkee), India in 1980
and 1994 respectivvely. Currently he is Professor in
Load Active Power 50 kW, t < 0.5, t > 0.8 the Department ofo Electrical Engineering, Indian
175 kW 0.5 < t < 0.8 Institute of Technology, Roorkee, Ind dia. His research areas include
electric machines and drives, power converters,
c power quality, active
filters and renewable energy systems. He
H has authored and co-authored
VIII. REFERENCES many technical papers in national and international journals and
conferences.
[1] F. Blaabjerg, Z. Chen, and S. B. Kjaer, ““Power electronics as
efficient interface in dispersed power generration systems,” IEEE R. D. Patidar was born
b in Mandsaur, MP, India in
Transactions on Power Electronics, vol. 19, no. 5, pp. 1184-1194, the year 1971. He graaduated in Electrical Engineering
Sep. 2004. from Govt. Engg. Co ollege, Rewa, India in 1995. He
[2] B. K. Bose, “Energy, environment, and advances in power completed his M.E E. in Power Electronics from
electronics,” IEEE Transactions on Power Ellectronics, vol. 15, no. SGSITS, Indore, Indiia in 2004 and Ph. D in the field
4, pp. 688-701, Jul. 2000. of Active Filters from
f Department of Electrical
[3] R. Teodorescu, F. Blaabjerg, M. Liserre, and A. V. Timbus, Engineering, Indian Institute
I of Technology, Roorkee,
“Overview of Control and Grid Synchronizzation for Distributed India in 2010. He woorked for two and half years with
Power Generation Systems,” IEEE Transaactions on Industrial Jindal Steel and Pow
wer Limited, India in the area of
Electronics, vol. 53, pp. 1398-1409, Oct. 20066. power generation. Hee has around 10 years of teaching
[4] W. Kramer, S. Chakraborty, B. Kroposkki, and H. Thomas, experience. Currently he is Professor in i the Department of Electrical
“Advanced Power Electronic Interfaces for Distributed Energy Engineering, Mandsaur Institute of Tecchnology, Mandsaur, India. His
Systems, Part-1, Systems and Topologiess,” Technical Report field of interest includes power quality y, active filters, and renewable
NREL/TP-581-42672, Mar. 2008. energy.
[5] L .P. Kunjumuhammed and Mahesh K. Mishra “A control
algorithm for single-phase active power ffilter under non-stiff
voltage source,” IEEE Trans. Power Electronn., vol. 21, no. 3, pp.
822-825, May 2006.
[6] A. Chandra, B. Singh, B. N. Singh, and K. Al-Haddad, “An
improved control algorithm of shunt activve filter for voltage
regulation, harmonic elimination, power-faactor correction, and
balancing of nonlinear loads,” IEEE Traansactions on Power
Electronics., vol. 15, no.3, pp: 495-507, 2000..
[7] B. Singh, K. Al-Haddad and A. Chandra, “A new control approach

202
View publication stats